Deep learning has been applied for motor imagery electroencephalogram(MI-EEG)classification in brain-computer system to help people who suffer from serious neuromotor disorders.The inefficiency network and data shorta...Deep learning has been applied for motor imagery electroencephalogram(MI-EEG)classification in brain-computer system to help people who suffer from serious neuromotor disorders.The inefficiency network and data shortage are the primary issues that the researchers face and need to solve.A novel MI-EEG classification method is proposed in this paper.A plain convolutional neural network(pCNN),which contains two convolution layers,is designed to extract the temporal-spatial information of MI-EEG,and a linear interpolation-based data augmentation(LIDA)method is introduced,by which any two unrepeated trials are randomly selected to generate a new data.Based on two publicly available brain-computer interface competition datasets,the experiments are conducted to confirm the structure of pCNN and optimize the parameters of pCNN and LIDA as well.The average classification accuracy values achieve 90.27%and 98.23%,and the average Kappa values are 0.805 and 0.965 respectively.The experiment results show the advantage of the proposed classification method in both accuracy and statistical consistency,compared with the existing methods.展开更多
Brain-computer interfaces (BCIs) records brain activity using electroencephalogram (EEG) headsets in the form of EEG signals;these signals can berecorded, processed and classified into different hand movements, which ...Brain-computer interfaces (BCIs) records brain activity using electroencephalogram (EEG) headsets in the form of EEG signals;these signals can berecorded, processed and classified into different hand movements, which can beused to control other IoT devices. Classification of hand movements will beone step closer to applying these algorithms in real-life situations using EEGheadsets. This paper uses different feature extraction techniques and sophisticatedmachine learning algorithms to classify hand movements from EEG brain signalsto control prosthetic hands for amputated persons. To achieve good classificationaccuracy, denoising and feature extraction of EEG signals is a significant step. Wesaw a considerable increase in all the machine learning models when the movingaverage filter was applied to the raw EEG data. Feature extraction techniques likea fast fourier transform (FFT) and continuous wave transform (CWT) were usedin this study;three types of features were extracted, i.e., FFT Features, CWTCoefficients and CWT scalogram images. We trained and compared differentmachine learning (ML) models like logistic regression, random forest, k-nearestneighbors (KNN), light gradient boosting machine (GBM) and XG boost onFFT and CWT features and deep learning (DL) models like VGG-16, DenseNet201 and ResNet50 trained on CWT scalogram images. XG Boost with FFTfeatures gave the maximum accuracy of 88%.展开更多
Signal drift and performance instability of brain-computer interface devices induced by the interface failure between rigid metal electrodes and soft human skin hinder the precise data acquisition of electroencephalog...Signal drift and performance instability of brain-computer interface devices induced by the interface failure between rigid metal electrodes and soft human skin hinder the precise data acquisition of electroencephalogram(EEG).Thus,it is desirable to achieve a robust interface for brain-computer interface devices.Here,a kind of polydopamine methacrylamide-polyacrylamide(PDMA-PAAM)hydrogel is developed.To improve the adhesion,dopamine is introduced into the polyacrylamide hydrogel,through the amino and catechol groups of dopamine in an organic-inorganic interface to build a covalent and non-covalent interaction.A strong attachment and an effective modulus transition system can be formed between the metal electrodes and human skin,so that the peeling force between the PDMAPAAM hydrogel and the porcine skin can reach 22 N m^(-1).In addition,the stable conductivity and long-term operating life of the PDMA-PAAM hydrogel for more than 60 days at room temperature are achieved by adding sodium chloride(NaCl)and glycerol,respectively.The PDMA-PAAM hydrogel membrane fabricated in this work is integrated onto a flexible Au electrode applied in a brain-computer interface.In comparison,the collected EEG signal intensity and waveform are consistent with that of the commercial counterparts.And obviously,the flexible electrode with PDMA-PAAM hydrogel membrane is demonstrated to enable a more stable and userfriendly interface.展开更多
基金Foundation item:the National Natural Science Foundation of China(Nos.62173010 and 11832003)。
文摘Deep learning has been applied for motor imagery electroencephalogram(MI-EEG)classification in brain-computer system to help people who suffer from serious neuromotor disorders.The inefficiency network and data shortage are the primary issues that the researchers face and need to solve.A novel MI-EEG classification method is proposed in this paper.A plain convolutional neural network(pCNN),which contains two convolution layers,is designed to extract the temporal-spatial information of MI-EEG,and a linear interpolation-based data augmentation(LIDA)method is introduced,by which any two unrepeated trials are randomly selected to generate a new data.Based on two publicly available brain-computer interface competition datasets,the experiments are conducted to confirm the structure of pCNN and optimize the parameters of pCNN and LIDA as well.The average classification accuracy values achieve 90.27%and 98.23%,and the average Kappa values are 0.805 and 0.965 respectively.The experiment results show the advantage of the proposed classification method in both accuracy and statistical consistency,compared with the existing methods.
文摘Brain-computer interfaces (BCIs) records brain activity using electroencephalogram (EEG) headsets in the form of EEG signals;these signals can berecorded, processed and classified into different hand movements, which can beused to control other IoT devices. Classification of hand movements will beone step closer to applying these algorithms in real-life situations using EEGheadsets. This paper uses different feature extraction techniques and sophisticatedmachine learning algorithms to classify hand movements from EEG brain signalsto control prosthetic hands for amputated persons. To achieve good classificationaccuracy, denoising and feature extraction of EEG signals is a significant step. Wesaw a considerable increase in all the machine learning models when the movingaverage filter was applied to the raw EEG data. Feature extraction techniques likea fast fourier transform (FFT) and continuous wave transform (CWT) were usedin this study;three types of features were extracted, i.e., FFT Features, CWTCoefficients and CWT scalogram images. We trained and compared differentmachine learning (ML) models like logistic regression, random forest, k-nearestneighbors (KNN), light gradient boosting machine (GBM) and XG boost onFFT and CWT features and deep learning (DL) models like VGG-16, DenseNet201 and ResNet50 trained on CWT scalogram images. XG Boost with FFTfeatures gave the maximum accuracy of 88%.
基金supported by the National Natural Science Foundation of China(U20A6001,11921002,and 11902292)Zhejiang Province Key Research and Development Project(2021C01183,2020C05004,and 2021C05007-4)the Natural Science Foundation of Zhejiang Province of China(LQ19E030003)。
文摘Signal drift and performance instability of brain-computer interface devices induced by the interface failure between rigid metal electrodes and soft human skin hinder the precise data acquisition of electroencephalogram(EEG).Thus,it is desirable to achieve a robust interface for brain-computer interface devices.Here,a kind of polydopamine methacrylamide-polyacrylamide(PDMA-PAAM)hydrogel is developed.To improve the adhesion,dopamine is introduced into the polyacrylamide hydrogel,through the amino and catechol groups of dopamine in an organic-inorganic interface to build a covalent and non-covalent interaction.A strong attachment and an effective modulus transition system can be formed between the metal electrodes and human skin,so that the peeling force between the PDMAPAAM hydrogel and the porcine skin can reach 22 N m^(-1).In addition,the stable conductivity and long-term operating life of the PDMA-PAAM hydrogel for more than 60 days at room temperature are achieved by adding sodium chloride(NaCl)and glycerol,respectively.The PDMA-PAAM hydrogel membrane fabricated in this work is integrated onto a flexible Au electrode applied in a brain-computer interface.In comparison,the collected EEG signal intensity and waveform are consistent with that of the commercial counterparts.And obviously,the flexible electrode with PDMA-PAAM hydrogel membrane is demonstrated to enable a more stable and userfriendly interface.
